MSGM Clinical Case Discussion August 2026
Updated: Sep 1
The Valproate Trap: When Parkinson’s" and Dementia Are Found in a
Pillbox
Prepared by Dr. Khaw Mae Jane
Supervised by Dato Dr. Tunku Muzafar Shah, Dr. Chee Sing Hui
Case Summary
Patient Profile Background
Mdm PCL is a 64-year-old Chinese lady (Clinical Frailty Scale 4) who was referred to the Geriatrics Clinic from a private neuropsychiatry practice. She lives with her 73-year-old husband in an urban condominium. She demonstrates excellent functional capacity, independently managing complex tasks including her personal finances (via online banking), a complicated medication regimen, and daily smartphone use (she abstains from driving solely due to fear).
Her past medical history is significant for left invasive ductal carcinoma (diagnosed in 2013, currently in remission after WLE, chemoradiotherapy, and five years of Tamoxifen) and cardiometabolic comorbidities (DM, HTN, DLP).
History of Presenting Illness
Mdm PCL has a long-standing history of bipolar mood disorder diagnosed in 2009. Past manic episodes were characterized by grandiosity and impulsivity, including attempting to purchase entire store inventories at KLIA in 2007 (requiring police intervention) and transferring large sums of money online under the delusion of hiring a celebrity for her 60th birthday. For over a decade, she was maintained on sodium valproate, lithium, and risperidone.
In April 2023, she developed hand tremors that severely interfered with feeding, alongside gait instability resulting in several non-injurious falls. Her neuropsychiatrist diagnosed her with Parkinson’s disease, discontinuing lithium and risperidone while up-titrating sodium valproate to 1000 mg daily. Concurrently, levodopa/benserazide and trihexyphenidyl were initiated. Following a suspected mood relapse in December 2023, quetiapine and lorazepam were added.
Subsequently, in 2024, her husband reported intermittent short-term memory lapses (e.g., occasionally forgetting a recent meal, misplacing items, and repetitive questioning). These prompted a diagnosis of dementia and the initiation of memantine, followed by donepezil. However, this reported cognitive decline appears inconsistent with her preserved instrumental activities of daily living (IADLs) and intact executive function.
Current Medications
Neuropsychiatric:
o Sodium Valproate ER 1000 mg OD
o Quetiapine 25 mg ON
o Lorazepam 0.5 mg ON
o Levodopa/Benserazide (Madopar) 125 mg TDS
o Trihexyphenidyl (Artane) 2 mg BD
o Memantine 5 mg OM
o Donepezil 5 mg OM
Cardiometabolic:
o Metformin 1 g BD
o Simvastatin 20 mg ON
Physical and Cognitive Examination
o General: Vitals were stable (BP 120/68, HR 71). Notably, she has a low body weight of 44.8 kg.
o Neurological Motor: Examination revealed an absence of mask-like facies, lead- pipe rigidity, or bradykinesia. She exhibited a prominent postural and kinetic tremor (particularly evident on the finger-to-nose maneuver) that suppressed completely when the limbs were supported at rest. During passive movement, this tremor superimposed on normal muscle tone, creating a false cogwheel-like sensation. Her gait demonstrated a mildly stooped posture with preserved arm swing and no delayed initiation or shuffling. However, she exhibited marked axial instability, prominent en bloc turning, and a propensity for retropulsion.
o Cognitive Mood: Her Montreal Cognitive Assessment (MoCA) score was 18/30, with pronounced deficits in visuospatial/executive function (1/5), attention (3/6), language (1/3), abstraction (1/2), and delayed recall (3/5). Her Geriatric Depression Scale (GDS) was 1/15.
Investigations (12/06/26)
o Therapeutic Drug Monitoring: Serum Valproate 609 µmol/L (Ref: 346-693 µmol/L).
o Hematology: Hb 10.3 g/dL, Platelets 149, Ferritin 23, TSAT 17%.
o Cardiometabolic: HbA1c 5.7%, FBS 3.9 mmol/L, LDL 1.8 mmol/L.
o Reversible Dementia Screen: Vitamin B12 304 pmol/L, Folate 36 nmol/L, TSH 3.02mIU/L, FT4 10.1 pmol/L.
o Renal Hepatic: Normal (Cr 47, Alb 40, AST 26, ALT 3).
o Neuroimaging: A baseline CT Brain is scheduled for 19/08/26.
Questions for Discussion
1. Given her low body weight (44.8 kg), how should we interpret her normal valproate level?
2. What prescribing cascade is evident in this case?
3. What differentiates her presentation from idiopathic Parkinson's disease?
4. What is the most appropriate next step in management, and how would you prioritize the deprescribing process?
We invite members to share their insights on these discussion questions in the comments section below. The formal case resolution will be posted at the end of the month.
Suggested Reading References
1. Hassamal S, Waller S, Reese K, Testa C. Reversible valproic acid-induced parkinsonism and cognitive impairment in an elderly patient with bipolar disorder
I. Turk Psikiyatri Derg. 2016;27(3):213-217.
2. Rochon PA, Gurwitz JH. The prescribing cascade revisited. Lancet. 2017;389(10081):1778-1780. doi:10.1016/S0140-6736(17)31188-1.
3. Thanvi B, Treadwell S. Drug induced parkinsonism: a common cause of parkinsonism in older people. Postgrad Med J. 2009;85(1004):322-326. doi:10.1136/pgmj.2008.073312
4. Sekiguchi K, Mashiko T, Koide R, et al. A case of long-term exposure to valproic
acid mimicking tremor-dominant parkinson's disease. Tremor Other Hyperkinet Mov (N Y). 2023;13:17. doi:10.5334/tohm.755.
5. Fox C, Richardson K, Maidment ID, et al. Anticholinergic medication use and
cognitive impairment in the older population: the medical research council cognitive function and ageing study. J Am Geriatr Soc. 2011;59(8):1477-1483.
doi:10.1111/j.1532-5415.2011.03491.x.
6. Masmoudi K, Gras-Champel V, Bonnet I, et al. Démence et troubles extra-
pyramidaux sous acide valproïque au long cours [Dementia and extrapyramidal
problems caused by long-term valproic acid]. Therapie. 2000;55(5):629-634.
7. Parsons C, Lim WY, Loy C, et al. Withdrawal or continuation of cholinesterase
inhibitors or memantine or both, in people with dementia. Cochrane Database Syst Rev. 2021;2(2):CD009081. doi:10.1002/14651858.CD009081.pub2.
Working Diagnosis Clinical Discussion
The primary working diagnosis is Valproate-Induced Parkinsonism and Dementia
(VIPD), 1 which triggered a severe and extensive prescribing cascade. 2
Several clinical clues strongly support this:
1. Pharmacokinetics weight disproportion: While her valproate level (609 µmol/L) is within the therapeutic reference range, this is a massive drug exposure for a 44.8 kg patient. This high tissue exposure correlates directly with her prominent kinetic/postural tremor and borderline thrombocytopenia.
2. The prescribing cascade polypharmacy: Her initial motor symptoms were likely triggered by the combined long-term exposure to sodium valproate, lithium, and risperidone. However, when lithium and risperidone were appropriately discontinued following her Parkinson diagnosis, 3 the cascade was paradoxically perpetuated by the up-titration of sodium valproate (likely intended as a compensatory measure to prevent bipolar relapse). This sustained the tremor drive 4 and resulted in the inappropriate addition of levodopa/benserazide and trihexyphenidyl. The subsequent addition of quetiapine established a highly deleterious pharmacological network:
Pharmacodynamic synergy: Concurrent GABAergic enhancement (valproate), H1- receptor blockade (quetiapine), and muscarinic receptor antagonism (trihexyphenidyl) synergistically amplified CNS depression.
Anticholinergic Cognitive Burden (ACB): This regimen yielded a cumulative ACB score of 7 (scores ≥ 3 are highly associated with cognitive impairment). 5
This iatrogenic cognitive blunting was misinterpreted as dementia, triggering yet another cascade with the initiation of memantine and donepezil.
3. Atypical motor cognitive phenotype: Several objective examination findings point away from idiopathic Parkinson’s disease:
Preserved arm swing.
Absence of lead-pipe rigidity or bradykinesia.
A prominent kinetic/postural tremor that suppresses completely when the limbs are supported at rest
Furthermore, her MoCA deficits (predominantly visuospatial and executive, attention and abstraction) reflect subcortical and frontal-executive processing delays typical of VIPD and anticholinergic burden, rather than primary neurodegenerative dementia. 6
Additional Clinical Considerations
Glycemic control: Investigations revealed excessively tight glycemic control (HbA1c 5.7%, FBS 3.9 mmol/L). In a vulnerable older adult with postural instability, this ypoglycemia risk is a major contributor to her fall, prompting an immediate reduction of her metformin to 500 mg BD.
Real-World Deprescribing Challenges
In this case, an initial clinical trial of tapering down trihexyphenidyl and levodopa/benserazide provoked a severe rebound tremor that impaired ambulation, prompting the patient to resume her previous doses to regain function. Interestingly, resuming the initial dose of trihexyphenidyl alone failed to control the rebound tremor; full suppression was only achieved when the previous dose of levodopa/benserazide was also reinstated.
This highlights a classic pharmacological trap:
Neuro-adaptation: The basal ganglia motor circuit relies on a delicate balance between dopaminergic and cholinergic signalling. Chronic valproate exposure severely suppressed her endogenous basal ganglia function, to the extent that she required the synergistic effects of both an anticholinergic (releasing the motor brake) and exogenous levodopa (driving motor output) to maintain baseline motor function.
Receptor downregulation: Chronic levodopa therapy likely downregulated her endogenous dopaminergic signalling, predisposing her to acute dopaminergic withdrawal upon dose reduction.
Cholinergic rebound: The dose reduction of trihexyphenidyl risks a cholinergic rebound due to muscarinic receptor hypersensitivity.
This confirms that deprescribing in a severe cascade requires targeting the root culprit before removing secondary symptom-masking agents.
Stepwise Deprescribing Plan
Address the root cause first: Liaise with psychiatry to carefully dose-reduce
valproate (adjusting for her low body weight) or cross-taper to an alternative mood stabilizer to diminish the underlying tremor drive.
Dementia agent de-escalation: Given the high probability that her cognitive deficits are iatrogenic, we will initiate a cautious withdrawal of memantine and donepezil. Discontinue one agent first, monitoring closely for any functional or cognitive decline. While evidence on the optimal withdrawal strategy is inconclusive, a gradual taper is preferred over abrupt cessation to mitigate potential withdrawal symptoms and patient distress. 7
Taper anticholinergics: As valproate levels decrease and tremor improves, slowly aper trihexyphenidyl to resolve the cumulative anticholinergic blockade.
Wean Parkinsonian agents: Gradually taper levodopa/benserazide. Although
ineffective for pure VIPD, her recent rebound tremor shows neuro-adaptation to exogenous dopamine, necessitating a gradual wean to avoid acute dopaminergic withdrawal.
Comments